Intended for healthcare professionals

Rapid response to:

Editorials

Molnupiravir’s authorisation was premature

BMJ 2022; 376 doi: https://doi.org/10.1136/bmj.o443 (Published 03 March 2022) Cite this as: BMJ 2022;376:o443

Rapid Response:

Concern about imbalance of baseline characteristics in molnupiravir trials

Dear Editor

The editorials entitled “Molnupiravir’s authorisation was premature” [1] emphasised the dangers of making decisions based on a single prematurely terminated trial. I would like to add some other serious problems in the MOVe-Out trial [2] and MOVe-IN trial [3] which reduce the robustness of efficacy and safety of molnupiravir that were not referred by the editorials.

(1) Sex-adjusted hazard ratio was not significant
First, MOVe-Out trial reportedf that hospitalisation or death (hospitalisation/death) in all-randomized population was not significant by the time-to-event analysis: hazard ratio (HR) was 0.69 (95%CI: 0.48-1.01) by adjusting the imbalance of sex [2].

(2) Molnupiravir may worsen COVID-19 after interim analysis
When calculating simply the proportion of hospitalization/death among population who were not included in the interim analysis but included in the final analysis, the molnupiravir group had a non-significant higher risk (6.2 %) than the placebo group (4.7%).

(3) Serious imbalance of baseline risk factors favouring molnupiravir
Move-Out trial [2] reported that baseline characteristics except sex were similar in the two groups including obesity (BMI≧30), age > 60 year (old age), diabetes mellitus (DM), chronic kidney disease (CKD), chronic obstructive pulmonary disease (COPD), serious heart disease, and active cancer. However, in the interim analysis population, proportion of participants with COPD was significantly less assigned in the molnupiravir group: odds ratio (OR) is 0.31 (95%CI: 0.13-0.73, p=0.0043) and those with obesity was marginally significantly more assigned in the molnupiravir group: OR=1.33 (95%CI:0.95-1.86, p=0.094) [4].
The sum of the percentages of the participants who have risk factors other than obesity was significantly lower in the molnupiravir group (43.4%) than in the placebo group (51.8%): OR=0.71 (95%CI:0.54-0.95, p=0.019). The sum of the percentages restricting 4 important risk factors (DM, CKD, COPD and active cancer), odds was almost 40% lower in the molnupiravir group: OR=0.61 (95%CI:0.43-0.85, p=0.0043) [4].

(4) Significant imbalances also in all-randomized population
Significant imbalances of baseline risk factors were also observed in the all-randomized population. The sum of the percentages having risk factors other than obesity was significantly lower in the molnupiravir group (53.8%) than in the placebo group (59.4%). OR=0.79 (95%CI:0.64-0.98, p=0.031). If the sum of the percentages restricting to the 4 important risk factors, OR=0.77 (95%CI:0.61-0.97, p = 0.026) [4].

(5) Doubt on fair random asignment and blinding
Significantly lower OR of risk factors in the analysis of all-randomized population raises doubt as to whether the randomization has been fairly conducted in the trial. A further significant imbalance in the interim analysis, and a great difference in COPD (OR=0.31, p=0.0043) suggests the possibility that before the interim analysis, blinding might have been broken.

(6) Apparent effect on participants with obesity is not conclusive
In a subgroup analysis, no significant reduction of hospitalization/death was observed in participants with diabetes nor with severe heart disease[2]. No data for efficacy was found for those with CKD, COPD nor with active cancer. Apparent significant reduction was reported only with obesity but it cannot be denied that it may be affected by the bias of other risk factors.

(7) Contradictory results in moderate to severe COVID-19
According to the document submitted to the UK regulator by Merck Sharp & Dohme (UK) Ltd [5], hospitalization/death was significantly reduced in the subgroup of moderate COVID-19 and the effect size was greater in the moderate patients (Risk difference=-8, 95%CI: -15.5, -0.5) than in the mild patients (-4.9, 95%CI:-10.5, 0.2).
However, two RCTs targetting moderate COVID-19 have been terminated due to futility by the interim analysis [6].

(8) MOVe-IN trial also has serious imbalance in baseline severity
In the MOVe-IN trial [3], participants with score 6 COVID-19 (hospitalized and administered oxygen by non-invasive ventilation or high flow) were significantly less assigned in molnupiravir groups (5/218=2.3%) than the placebo group (6/75=8.0%): OR=0.27 (95%CI:0.08-0.91, p=0.025). Dispite this serious bias favouring molnupiravir group, non-significant increase of death was reported in molnupiravir group (13/218=6.0%) compared with placebo group (2/75=2.7%): OR=4.69 (95%CI: 0.60-36.50, p=0.105). If the baseline imbalance favourring molnupiravir group is adjusted, OR for all-cause mortality may be statistically significant. For example, the ratio of mortality odds ratio to odds ratio of baseline score 6 was 17.38 (95%CI:1.60-188.84) according the methods described by Stephan Kalossa [7].

(9) Reanalysis is needed and the reasons of serious imbalance should be verified.
It is necessary to reanalyse acurate incidence of hospitalization or death (MOVe-Out trial) and mortality rate (MOVe-IN trial) by adjusting baseline risk factors (MOVe-Out trial) and baseline severity especially of score 6 (MOVe-IN trial). Moreover, it shoudl be verified why these serious imbalance of baseline characteristics occured in the molnupiravir trials.

(10) Beware of DNA damage, bone marrow toxicity and mutations
Molnupiravir is a substance similar to some antivirals including favipiravir or ribavirin, which causes mutations not only in viral genes to prevent proliferation but also which may suppress human cell division [8-11]. Irreversible myelosuppression was observed in dogs treated with 0.4-fold the human equivalent dose according to the AUC level of active form of molnupiravir at the 800 mg q 12h human dose for 22 days and NOAEL (No-observed-adverse-effect level) is only 0.13-fold the human dose [5].
It shoud be clarified whether more death observed in the MOVe-IN trial [3] is associated with bone marrow toxicities.

(11) Full clinical study reports should be disclosed and reanalysis is needed
As was warned by Sidebottom et al in the editorial of BMJ, in the absence of sufficient evidence of safety and efficacy raises serious concerns about further mistakes being made [12]. It seems that their warnings have come true. The RCTs of antiviral agents for SARS-CoV-2, such as remdesivir [13,14] and baricitinib [15] also have various contradictions.
In order to resolve these situations, as was conducted in the systematic reviews on neuraminidase inhibitors [16] including oseltamivir (Tamiflu) [17.18], clinical study reports should be disclosed and should be examined.

Conflict of interest: none

References
1. Brophy, JM. Molnupiravir’s authorisation was premature. BMJ 2022;376:o443 https://www.bmj.com/content/376/bmj.o443
2. Jayk Bernal A, Gomes da Silva MM, Musungaie DB et al. (MOVe-OUT Study Group) Molnupiravir for Oral Treatment of Covid-19 in Nonhospitalized Patients. N Engl J Med. 2022 Feb 10;386(6):509-520 (2021 Dec 16 Online ahead of print. With supplementary appendix. PMID: 34914868
3. Arribas JR, Bhagani S, Lobo SM et al. Randomized trial of molnupiravir or placebo in patients hospitalized with covid-19. NEJM Evidence2022;1:EVIDoa2100044. doi:10.1056/EVIDoa2100044 .
4. MedCheck Editorial team. Antiviral pill for COVID-19 Molnupiravir (Lagevrio ?) : No confirmed efficacy, No efficacy for high-risk people such as diabetes. Web MedCheck No 202 (January 28, 2022) in Japanese:
https://www.npojip.org/sokuho/220128.html
5. Merck Sharp & Dohme (UK) Ltd. Public Assessment Report National Procedure Lagevrio 200mg hard capsules (molnupiravir)
https://assets.publishing.service.gov.uk/government/uploads/system/uploa...
6. Singh AK, Singh A, Singh R et al. Molnupiravir in COVID-19: A systematic review of literature. Diabetes Metab Syndr. 2021 Nov-Dec;15(6):102329. doi: 10.1016/j.dsx.2021.102329. Epub 2021 Oct 30.PMID: 34742052
7. Kolassa S. Calculate 95% CI for ratio of odds ratio. https://stats.stackexchange.com/questions/471805/calculate-95-ci-for-rat...
8. van Schalkwyk JM Buyer beware: molnupiravir may damage DNA. .BMJ. 2021 Nov 4;375:n2663. doi: 10.1136/bmj.n2663.PMID: 34737196
9. Kabinger F, Stiller C, Schmitzová J, et al. Mechanism of molnupiravir-induced SARS-CoV-2 mutagenesis. Nat Struct Mol Biol2021;28:740-6. . doi:10.1038/s41594-021-00651-0 PMID:34381216
10. Zhou S, Hill CS, Sarkar S et al. β-d-N4-hydroxycytidine inhibits SARS-CoV-2 through lethal mutagenesis but ss also mutagenic to mammalian cells. J Infect Dis2021;224:415-9. doi:10.1093/infdis/jiab247 PMID:33961695
11. Gordon CJ, Tchesnokov EP, Schinazi RF, Götte M. Molnupiravir promotes SARS-CoV-2 mutagenesis via the RNA template. J Biol Chem. 2021 Jul;297(1):100770. doi: 10.1016/j.jbc.2021.100770. Epub 2021 May 11.PMID: 33989635
12) Sidebottom DB, Smith DD, Gill D. Safety and efficacy of antivirals against SARS-CoV-2 BMJ 2021; 375
doi: https://doi.org/10.1136/bmj.n2611 (Published 28 October 2021)
13) Med Check editorial team, Remdesivir (trade name: Veklury) Most likely ineffective for COVID-19. MedCheck in English 2020: 6 (18): 39-45. https://www.npojip.org/english/MedCheck/Med%20Check%20Tip-18-2020-08-21.pdf
With supplementary appendix at: https://www.npojip.org/english/MedCheck/Remdesivir%20Supplementary%20Sli...
14) Med Check editorial team, Remdesivir: No efficacy by WHO’s RCT. Med Check in Japanese2021:21(93):13-14 (in Japanese). Available at: https://www.npojip.org/chk_tip/No93-f05.pdf
15) Med Check editorial team, Baricitinib for the treatment of COVID-19: Data are seriously inconsistent and unreliable. 2021: 21 (96): 82-85. (in Japanese) https://www.npojip.org/chk_tip/No96-f05.pdf
16) Jefferson T, Jones MA, Doshi P, et al. Neuraminidase inhibitors for preventing and treating influenza in healthy adults and children. Cochrane Database Syst Rev 2014 Apr 10;2014(4):CD008965. Doi: 10.1002/14651858.CD008965.pub4.
17) Hama R, Bennett CL The mechanisms of sudden-onset type adverse reactions to oseltamivir. Acta Neurol Scand. 2017 Feb;135(2):148-160. doi: 10.1111/ane.12629. Epub 2016 Jun 30. PMID: 27364959
18) Hama R. The mechanisms of delayed onset type adverse reactions to oseltamivir. Infect Dis (Lond). 2016 Sep;48(9):651-60. doi: 10.1080/23744235.2016.1189592. Epub 2016 Jun 2.PMID: 27251370

Competing interests: No competing interests

07 March 2022
Rokuro Hama
Medical Doctor
NPO Japan Institute of Pharmacovigilance
#602 Katsuyama 2-14-8, Tennoji-ku, Osaka, Japan 543-0043